Most people use AI for analysis once. They prompt it, read the answer, and move on. That is not analysis. That is a guess with extra steps.
The value is not the single answer. It is the loop you run around it.
Five steps, AI builds and you judge:
-> ASK: write the actual question. "Is this a good company" is a vibe. "Did gross margin compress two years running, and why" is a question AI can work. -> RETRIEVE: make it pull the evidence with a source on every line. The 10-K, the earnings call, the comps. No source means the claim does not exist yet. -> VERIFY: every figure gets a source, a page, and a date, or it gets cut. You check the ones that move the thesis against the primary filing yourself. -> CHALLENGE: ask AI to build the strongest case against your read. If it cannot, you have not stress-tested it. You have just agreed with yourself. -> MONITOR: the loop does not close. New filing, new call, new print. Re-run the steps when the facts change.
Notice what AI never does here. It does not decide if the business is good, what it is worth, or whether you act. That is your assumption set, not its output.
The analyst who wins is not the one with the best prompt. It is the one who runs the loop with discipline.
Save this and run it on your next name.
Educational content only. Not investment advice, and not a recommendation to buy, sell, or hold any security. Wall Street Prompt. Always verify against the primary source.